codegen-diagram

基于当前项目/代码生成 Draw.io 图表，支持技术栈图、系统架构图、数据结构图、E-R 图四种类型。输出符合 Draw.io 语法的 .drawio 文件（mxGraph XML），可直接导入 Draw.io 编辑。当用户提到技术栈、系统架构、数据结构、E-R 图时使用。

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is well-structured and concise with a clear task-routing table and concrete style specs, but it lacks an executable mxGraph XML template and any validation step for the fragile XML output. The four referenced reference files are also missing, breaking the intended progressive-disclosure navigation.

Suggestions

Add a minimal, copy-paste mxGraphModel XML scaffold (root + one node + one edge) so the agent has an executable starting point rather than only output requirements.

Include a validation checkpoint in the output workflow (e.g. "verify the XML is well-formed and imports into Draw.io before finalizing") to satisfy the feedback-loop expectation for XML/document operations.

Provide the four referenced reference/ files (tech-stack.md, system-arch.md, data-structure.md, er-diagram.md) or remove the dead links so progressive-disclosure navigation actually resolves.

Dimension	Reasoning	Score
Conciseness	Lean and token-efficient: uses compact tables for keyword routing and color specs with no padding or explanation of concepts Claude already knows.	3 / 3
Actionability	Provides concrete values (colors #3366CC, font Helvetica 11–13, arrow style endArrow=classicBlockThin) but no copy-paste mxGraph XML scaffold or executable example; "完整 mxGraph XML" is stated as an output requirement rather than demonstrated.	2 / 3
Workflow Clarity	A keyword→reference routing table and a 4-step output list give a clear sequence, but there is no validation checkpoint (e.g. verify XML well-formedness / Draw.io importability) for the fragile XML-generation operation, capping the score per the rubric's feedback-loop rule.	2 / 3
Progressive Disclosure	Body points to four one-level-deep reference files (reference/tech-stack.md etc.) with clear signaling, but the referenced files do not exist in references/ (the directory is empty), so navigation is broken rather than well-organized.	2 / 3
	Total	9 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is concise, specific, and complete: it states concrete capabilities for four diagram types, provides a .drawio output format, and includes an explicit "Use when..." trigger clause. It is distinct from other skills and avoids verbosity.

Dimension	Reasoning	Score
Specificity	Names concrete actions ("生成 Draw.io 图表", "输出...符合 Draw.io 语法的 .drawio 文件（mxGraph XML）", "可直接导入 Draw.io 编辑") across four distinct diagram types.	3 / 3
Completeness	Explicitly states both what it does (generate four Draw.io diagram types as .drawio files) and when to use it ("当用户提到...时使用").	3 / 3
Trigger Term Quality	Includes natural user-facing keywords via "当用户提到技术栈、系统架构、数据结构、E-R 图时使用", covering common phrasings per chart type.	3 / 3
Distinctiveness Conflict Risk	Clearly scoped to project/code-based Draw.io diagrams with distinct triggers, making conflicts with other skills unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: xstongxue/best-skills
Path: skills/codegen-diagram/SKILL.md
Commit: fd28854

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.